LUCIA AND ATREVIDA GLACIERS 67
the eastern margin of Terrace Point, forming a lake which did not exist in 1905. This lake still existed August 3,1909, but south of it for over half a mile the moraine-covered depression between Lucia and Atrevida Glaciers was covered with icebergs. The expanded lake in which these icebergs floated was seen from the crest of Amphitheatre Knob on July 21, 1909, but before August 3 was drained through a subglacial tunnel.
Another notable change along the eastern margin of the Lucia Glacier was the development of a huge glacial stream in the depression between the outer portions of the Lucia and Atrevida lobes. In 1906 there was only a small stream here, but in 1909 one of the largest glacial streams of the region was flowing. This increase in volume was probably due in part to the recent breaking of the glacier, thus exposing the ice to more rapid melting; but this does not seem to be a sufficient cause for so great an increase. It is probable that a still more efficient cause is the interference with and destruction of the subglacial drainage which had developed during years of stagnation, and which in 1905 and 1906 found escape at the southern end of Lucia Nunatak on the opposite, or western, side of the glacier. The advance and breaking of the glacier, so notable near the nunatak, must have destroyed this system of subglacial drainage toward which most of the water from the glacier formerly flowed. Whether the new drainage all went to the east side, or whether a considerable portion still emerged on the western margin could not be determined; but that a large proportion escaped through the eastern stream is certain.
This remarkable change in condition of Lucia Glacier in the interval between August, 1906, and August, 1909, had wholly altered the appearance of the glacier, and yet we conclude that the thrust by which the change has been caused was either a weak one or else only just beginning to make its efforts felt, probably the latter. The reasons for this conclusion are several, all based upon the changes which a similar advance brought about in other glaciers. Compared with the Atrevida in 1906, for instance, the Lucia hi 1909 was far less broken, and the piedmont area was not noticeably thickened, as the Atrevida was. As yet there had been no development of great crescentic crevasses in the piedmont bulb as was the case in the Atrevida, and the crevassing did not reach out into the alder zone as it did in the Atrevida. In other words, the stage of breaking of Lucia Glacier was in 1909 far less advanced than that of Atrevida Glacier in 1906. It was far less broken than the Marvine, Haenke and Variegated Glaciers also.
In view of the rapidity with which these waves of advance pass down a glacier, essentially the whole transformation hi the case of those that advanced in 1906 having taken place within a period of nine or ten months, it is not improbable that the lesser development of the Lucia advance was not due to weakness of the thrust but to the fact that our observations were made too early and before the advance was finished. We can state with certainty that the entire breakage was recent, and mainly, if not entirely, in the season of 1909, and that it was still in progress in August. The proof of this is conclusive and of several kinds. Most noticeable of all is the condition of the crevasses and inter-crevasse or serac areas in the middle of the glacier. Although well below snow-line, and in a situation where ablation is rapid, the edges of the crevasses were sharp and angular as if freshly-broken, and the serac areas were still so completely coverepl with ablation moraine except on the crevasse faces that little ice was seen in this entire area. Even near the glacier margin, where the ice was so broken and pinnacled, a large part of the